A central aim of neuroscience is to map neural circuits in the mammalian brain, in order to learn how they account for mental activities and behaviors, and how their alterations lead to neurological and psychiatric disorders. Recently, we developed Brainbow transgenic mice, which facilitate tracing of such circuits. By using the Cre-lox recombination system in a novel way, they color individual neurons in both central and peripheral nervous systems in one of-100 distinct hues. Numerous investigators plan to use Brainbow lines, now in a public repository, for tracing neuronal connectivity in normal mice and in mouse models of human disease. Nonetheless, the Brainbow method is presently imperfect in several respects. We therefore propose to generate, evaluate and distribute second-generation Brainbow transgenic lines that circumvent limitations of the first generation. (1) To expand the range of cell types and developmental stages to which Brainbow methods can be applied, we will generate new lines using other regulatory elements. (2) To expand the spectrum and overcome photoinstability, we will test new fluorescent proteins, as well as an epitope tagging strategy that will allow Brainbow to span the spectrum from deep blue to far red. (3) In the Brainbow lines generated so far, colors are used to distinguish one neuron for another, but color choice is totally arbitrary. We will generate new lines in which colors serve to identify neuronal type as well as to distinguish individual neurons within a type. (4) Using new proteins and lox combinations, we will generate new lines in which many 100s of hues are expressed. (5) We will generate lines in which both axonal and dendritic arbors are fully labeled, and also generate lines in which synaptic sites are highlighted. Over the course of 4 years, we propose to generate and test approximately 200 transgenic lines (5 from each of 40 constructs). We believe this program will circumvent many if not all of the limitations enumerated above. We will make all results (both positive and negative) and constructs publicly available, and deposit constructs the most promising lines in a public repositories, so they can be broadly distributed.